This proposal will gather data relevant to the mechanisms by which widely used, non-drug treatments of headache (HA) have their benefits. While the long-term goal of this program of research is to compare three different models of the mechanisms by which non-drug treatments (thermal biofeedback [TBF] and progressive muscle relaxation [PMR] lead to the alleviation of chronic HA, the present proposal is designed to gather preliminary data to assist in the design of later, large scale, more definitive studies. The three models are the older psychophysiological model which posits that treatment leads to physiological change which in turn leads to HA relief. A new cognitive-attributional model posits that treatment leads to cognitive or attributional changes which lead re-interpretation of stressful, headache-causing events, which in turn leads to HA relief. A third interactional model posits that it is the interaction of the psychophysiological effects and cognitive- attributional effects which lead to HA relief. In Study 1 of this proposal, four different biofeedback conditions will be compared in the treatment of vascular HA: (1) TBF for hand warming, the most widely used non-drug treatment for vascular HA; (2) TBF for hand cooling, which some studies have shown is equally effective as (1); (3) TBF for hand temperature stabilization, which some claim is central to TBF's therapeutic effects; and (4) EEG alpha biofeedback for alpha suppression, to control for any vascular effects. We will also examine whether patients with pure migraine HA and those with mixed migraine and tension HA respond in the same way to the treatments. In Study 2 tension HA patients will receive PMR with computerized end of session feedback on degree of relaxation (to manipulate attributions of success) in order to have the equivalent of 2X2 design, two levels of cognitive-attributional change by varying degrees of physiological change. The results and models will be evaluated both with headache outcome measures (relative degree of headache relief) and with process measures evaluating psychophysiological change, cognitive and attributional change, and change in life stresses, mood and coping as they all relate to headache relief.